CSC 8980
Deep Reinforcement Learning
|
[1/10/23] Please check announcements from iCollege.
[1/10/23] Please ask questions at Course Discussion List.
Instructor: Jonathan Shihao Ji |
TA: TBD |
Office: One Park Place, Room 637 |
Office: One Park Place, Room 625 |
Office Hour: Mon. 3-4pm |
Office Hours: TT 3-4pm |
Email: TBD |
Email: TBD |
Lecture location: Langdale Hall 615 Lecture times: TT 12:45 - 2:30 PM
About the Course
This course introduces the concepts and algorithms of Deep Reinforcement Learning (DRL). It covers Markov decision process (MDP) and its tabular solvers such as policy iteration, value iteration, SARSA, Q-learning, as well as the modern deep neural network based RL solvers, such as DQN, Policy Gradient, Actor-Critic. Function approximators including MLP, CNN, LSTM, Transformer will be presented. Depending on the course progess, selected topics such as AlphaGo, AlphaGo Zero, and Inverse Reinforcement Learning will be discussed. The class emphasizes on the understanding of RL algorithms as well as their practical implementations with Python.
Textbook
Reinforcement Learning: An Introduction, Richard S. Sutton and Andrew G. Barto, MIT Press, 2018
Prerequisites
- This should not be your first machine learning class! You should have taken CSC 4740 Data Mining or CSC 4980 Machine Learning with a B or above.
- Basic knowledge in statistics, linear algebra, and calculus
- Proficiency in Python: All HWs will be in Python
Grading
Participation |
10% |
Assignments |
40% |
Final Exam |
20% |
Project |
30% |
|
|
A+ [97, 100]
| A [93, 97)
| A- [90, 93)
|
B+ [87, 90) |
B [83, 87) |
B- [80, 83) |
C+ [75, 80) |
C [70, 75) |
D [60, 70) |
F [0, 60) |
|
- No late submissions accepted! All the assignment and project deadlines are 11:59 pm (EST) of the due dates.
- If you have doubts in your grading, please email TA and CC to the Instructor indicating the reason why you think it should be regraded.
- The regrading request should be submitted within 1 week after you receive your score.
Misc
- Course materials, assignments, projects and QAs are managed by iCollege.
Academic Honesty Policy
All work submitted for grading must be the student’s own work. A student who submits an assignment that copies the work of another student,
in whole or in part, will be assigned a grade of zero for that assignment. Any student found to be cheating on an examination will receive a
score of zero for that exam. Cheating on an assignment or exam may result in dismissal from the course and notification of the Dean of Students.
More details on academic honesty can be found here.
|